Gut, 1976, 17, 797-800
Arterial blood gas tensions during upper gastrointestinal endoscopy P. J. WHORWELL, C. L. SMITH, AND K. J. FOSTER From the Professorial Medical Unit, Southampton University Hospitals, Southampton
Arterial blood gas tensions were measured before and during upper gastrointestinal endoscopy, with (group 1) and without (group 2) sedation with intravenous diazepam. There was a highly significant fall in the PaO2, which occurred in both groups and was therefore not attributable to diazepam. Measurement of FEV1 and FVC before endoscopy had no predictive value for those patients whose PaO2 fell the most.
SUMMARY
Upper gastrointestinal endoscopy is a safe procedure with a low morbidity and mortality. With its increasing use, previously described complications (Schiller and Prout, 1976), of which those of a respiratory nature are most prominent, could become more apparent. Cyanosis is not infrequently noted during the examination, respiratory depression or apnoea may result from the sedatives administered
upper gastrointestinal endoscopy and gave fully informed consent for the study were investigated. One hour before the procedure, the FEV1 and FVC were measured, using a wedge bellows spirometer (Vitalograph Ltd.). Symptoms and past history of cardiovascular disease, respiratory disease, and' smoking habit were recorded. The results of tho previous chest radiograph and ECG were noted. Fifty patients (group 1) received 15 puffs of metered lignocaine throat spray (Xylocaine spray, Astra Chemicals Ltd.: one puff equals 10 mg lignocaine), followed by intravenous sedation with diazepam. A further 15 patients (group 2) were subjected to endoscopy, having only 15 puffs of throat spray and no premedication of intravenous sedation. A radial arterial puncture was made for a baseline arterial blood gas tension analysis, after which the patient's throat was sprayed with lignocaine and in group 1 intravenous diazepam administered until drowiness was achieved, the amount; given being noted. Endoscopy was performed with an, Olympus GIFK endoscope and a second arterial' blood sample was taken when the stomach and' duodenum had been examined, but with the endo-scope still remaining in situ. This occurred at a mean time of 12 minutes. Blood gases were measured using. a Radiometer ABL 1 automatic blood gas analyser. The patients were monitored electrocardio-graphically throughout the procedure and any arrhythmias noted. In addition, the efficacy of' sedation and the occurrence of cyanosis were noted.
(Dundee and Haslett, 1970), and pulmonary aspiration can occur in up to 29 % of procedures (Prout and Metreweli, 1972). Electrocardiographic abnormalities have been demonstrated in a variable proportion of examinations (De Masi and Akdamar, 1969; Sturges and Krone, 1973; Py6rala et a!., 1973; Fujita and Kumura, 1975), particularly in the presence of ischaemic heart disease, and it has been suggested that hypoxia may be a contributory factor. Although blood gas tensions have been shown to change during fibreoptic bronchoscopy (Salisbury et al., 1975), to our knowledge such changes have not been studied during upper gastrointestinal endoscopy. The purpose of this study was to measure the change, if any, in arterial blood gas tensions before and during endoscopy and to relate any changes observed to the administration of sedative and occurrence of cardiac arrhythmias. Forced expiratory volume in one second (FEV1) and forced vital capacity (FVC) were measured beforehand to assess their predictive value for patients at risk. Methods
Results
Sixty-five consecutive patients who presented for Received for publication 29 June 1976
Table 1 gives the details of the patients in the studyand Table 2 shows the results for PaO2 and PaCO2,. 797
798
P. J. Whorwell, C. L. Smith, and K. J. Foster
Table 1 Details ofpatients Group Male
Female
no. % no.
1
I1I
35 70 15 11 73 4
Y/.
30 27
Age(yr) Mean
Mean Hb Range
(yr) 55 6 55 3
18-82
26-76
(gldl)
12-8 13-4
Diagnosis Ulcer
Gastritis
Gast. no. %
Duod. no. Y.
no.
7 3
4 1
8 3
14 20
8 7
Duodenitis Carcinoma Normal
Other
Y.
no.
16
7 3
20
% 14 20
no.
%
no.
%
no.
%
4 0
8
17 4
34
26
3 1
6 7
Table 2 Arterial PaO2 and PaCO2 before and during endoscopy with and without diazepam Group
No.
Initial PaO, (KPa)
PaOQ during endoscopy (KPa)
Percentage difference
Mean ± SE
Mean i SE
Mean i SE
10-62
I
50
12-32 ± 0-24
(Diazepam) II (No diazepam)
15
12-34
1
49
Initial PaCO, 4-87 i 0-08
15
4-63 i 011
0 53
0-23
13-39
1-42*
10-36 t 0-42
14-78
3-69*
PaCO2 during endoscopy
Percentage difference 7-08 1-40*
520 i 009
(Diazepam) II (No diazepam)
4-49
0-13
3-63
2-28
*P < 0-001.
obtained before and during the examination in both groups of subjects. In group 1 the mean PaO2 of 10-6 + 0-23 KPa found during the examination is significantly lower than the mean initial PaO2 of 12-3 + 0-24 KPa (p < 0-001). The mean percentage difference in PaO2 between the initial and subsequent readings of 13-4% is highly significant (p < 0-001). Similar results for PaO2 were obtained in group 2, where no diazepam was used. The number of patients examined in this group was limited to 15 as the study was continued only until it had been confirmed statistically that there was no difference, in respect of the fall of PaO2, between the two groups. In group 1 there was a significant increase in the PaCO2 during endoscopy, contrasting with a marginal decrease in PaCO2 in group 2 where no diazepam was administered. The Figure shows the least squares regression line for both groups for the correlation between the percentage change in PaO2 and the starting value. There was no significant difference in the separate regression lines for the two groups, and thus the combined regression line, which is highly significant (p < 0-01), is shown. It is clear that in all but two subjects there was a fall from the initial PaO2, regardless of the starting value, although there was a tendency for those with the highest values to fall by a greater extent. Although the FEV1 correlated with the initial PaO2 (P < 0-05),
neither the FEV1 nor the FVC were of predictive value for the percentage fall in PaO2. After the exclusion of those patients with preexisting arrhythmias, in group 1 arrhythmias occurred in seven patients (14 %), of whom two gave a history of ischaemic heart disease. In group 2, only one patient, who had ischaemic heart disease, had an arrhythmia. All arrhythmias except one were multiple ventricular ectopics, the exception being atrial fibrillation. All reverted spontaneously after the procedure. The occurrence of arrhythmias was not related to the degree of fall in the PaO2. There was no correlation between the dose of diazepam administered or depth of sedation produced and the percentage fall in oxygen tension. Discussion
This study clearly shows that there is a marked fall in the arterial PaO2 during uncomplicated upper gastrointestinal endoscopy. Although the mean PaO2 during endoscopy was only just below the lower limit of normal, in many subjects it fell significantly further, the lowest recorded value being 5-9 KPa. The observation that there was an equally notable fall in PaO2 when no diazepam was administered indicates that this effect is not due to the diazepam. It therefore seems probable that it is either the presence
Arterial blood gas tensions during upper gastrointestinal endoscopy * With
0to °2 Change 50
799
Diazepam
* Without Diazepam
40 _
30
a
Figure The percentage change in PaO2 plotted against the starting value, with the least squares t regression line for the combined data, is shown. The curved lines show the 95 % confidence limits for the regression line.
a
20
10
8 -10
10
12
14
*
16
18 Before 02
Endoscopy
-20
of the endoscope or the lignocaine throat spray, or a References combination of the two, which is responsible for this finding. It has been previously reported that ligno- De Masi, C. J., and Akdamar, K. (1969). Electrocardicgraphy during upper gastrointestinal endoscopy. Gastrointestinal caine throat spray may affect blood gas tensions Endoscopy, 16,33-34. (Salisbury et al., 1975). However, diazepam may Dundee, J. W., and Haslett, W. H. K. (1970). The benzocontribute to some extent in that the mean PaCO2 diazepines. British Journal of Anaesthetics, 42, 217-234. rose only when diazepam was administered. Dunn, G. D., Kubin, R. H., Laing, R. R., Sisk, C. W., and Klatz, A. P. (1970). Double-blind study of endoscopy preAlthough hypoxia may contribute to the generamedications. Gastrointestinal Endoscopy, 16, 229-230. tion of cardiac arrhythmias, in this study no clear Fujita, R., and Kumura, F. (1975). Arrythmias and ischemic relationship between hypoxia and cardiac rhythm changes of the heart induced by gastric endoscopic abnormalities could be shown in our patients. procedures. American Journal of Gastroenterology, 64, 4448. Although upper gastrointestinal endoscopy is a G. R., Kehoe, E. L., Friedman, E., and Belber, J. safe and diagnostically invaluable procedure, the Mayes, (1970). Pre-endoscopic medication: parenteral diazepam present study indicates that circumspection should used adjunctively. Gastrointestinal Endoscopy, 16, 187-193. be used in submitting to endoscopy those patients Prout, B. J., and Metreweli, C. (1972). Pulmonary aspiration after fibre-endoscopy of the upper gastrointestinal tract.. with respiratory disease, and the previously advoBritish Medical Journal, 4, 269-271. cated use of potential respiratory depressants as pre- Prout, B. J., and Schiller, K. F. R. (1976). In Modern Topics medication (Dunn et al., 1970; Mayes et al., 1970; in Gastrointestinal Endoscopy, pp. 73-81. Edited by K. F. R. Prout and Schiller, 1976) is to be discouraged. Schiller and P. R. Salmon. Heinemann: London. We are grateful to Mr D. Carpenter for the blood gas analyses and to Mr J. Alexander for his assistance with the statistical analysis.
Pyorala, K., Salmi, H. J., Jussila, J., and Heikkila, J. (1973).. Electrocardiographic changes during gastroscopy. Endoscopy, 5, 186-193. Salisbury, B. G., Metzger, L. F., Altose, M. D., Stanley, N. N., and Cherniack, N. S. (1975). Effect of fibreoptic
P. J. Whorwell, C. L. Smith, and K. J. Foster
800 bronchoscopy on respiratory performance in patients with chronic airways obstruction Thorax, 30, 441-445. Schiller, K. F. R., and Prout, B. J. (1976). In Modern Topics in Gastrointestinal Endoscopy, pp. 147-165. Edited by K. F. R. Schiller and P. R. Salmon. Heinemann: London. Sturges, H. F., and Krone, C. L. (1973). Cardiovascular stress
of peroral gastrointestinal endoscopy. Gastrointestinal Endoscopy, 19, 119-122. Requests for reprints: Dr. C. L. Smith, Professorial Medical Unit, Southampton General Hospital, Tremona Road, Southampton S09 4XY.
The September 1976 Issue THE SEPTEMBER 1976 ISSUE CONTAINS THE FOLLOWING PAPERS
Histopathology of cell mediated immune reaction in mouse colon-allograft rejection R. J. HOLDEN AND ANNE FERGUSON
Ischaemic colitis in the experimental animal I Comparison of the effects of acute and subacute vascular occlusion
J. G. W. MAITHEWS AND T. G. PARKS
Ischaemic colitis in the experimental animal U Role of hypovolaemia in the production of the disease
Idiopathic cholestasis of pregnancy in a large kindred
HUMBERTO REYES, J. RIBALTA, AND M. GONZALEZ-CERON
Autoimmune reaction to a liver specific membrane antigen during acute viral hepatitis A. M. G. COCHRANE, A. MOUSSOUROS, A. SMITH, A. D. THOMSON, A. L. W. F. EDDLESTON, AND ROGER WILLIAMS
Gastritis duodenitis, and circulating levels of gastrin in duodenal ulcer before and after vagotomy D. D.
J. G. W. MATrHEWS AND T. G. PARKS
MEIKLE, K. B. TAYLOR, S. C. TRUELOVE, AND R. WHITEHEAD
Exocrine pancreatic function in juvenile-onset dia-
Tranexamic acid and upper gastrointestinal haemorrhage-a double blind-trial J. C. BIGGS, T. B. HUGH,
betes mellitus B. M. FRIER, J. H. B. SAUNDERS, K. G. WORMSLEY, AND I. A. D. BOUCHIER
Stenosis of the colon in acute pancreatitis w. s. J. MAIR, M. J. MCMAHON, AND J. C. GOLIGHER
AND A. J. DODDS
Transient paraproteinaemia in a patient with coeliac disease A. S. PENA, JANNY V. NIEUWKOOP, H. R. E. SCHUIT, W. TH. J. M. HEKKENS, AND A. J. CH. HAEX
Technique Peroral small-intestinal biopsy: experience Ferritin in Crohn's disease tissue: detection by with the hydraulic multiple biopsy instrument in electron microscopy P. J. WHORWELL, R. C. routine clinical practice BRIAN B. scolT AND M. S. BALDWIN, AND R. WRIGHT
LOSOWSKY
Hypergastrinaemia in cirrhosis of liver sHu KUM
Notes and activities Notes on books
LAM
Copies are still available and may be obtained from the PUBLISHING MANAGER, BRMSH MEDICAL ASSOCIATION, TAVISTOCK SQUARE, LONDON, WC1H 9JR, price £2-00, including postage
Arterial blood gas tensions during upper gastrointestinal endoscopy P. J. WHORWELL, C. L. SMITH, AND K. J. FOSTER From the Professorial Medical Unit, Southampton University Hospitals, Southampton
Arterial blood gas tensions were measured before and during upper gastrointestinal endoscopy, with (group 1) and without (group 2) sedation with intravenous diazepam. There was a highly significant fall in the PaO2, which occurred in both groups and was therefore not attributable to diazepam. Measurement of FEV1 and FVC before endoscopy had no predictive value for those patients whose PaO2 fell the most.
SUMMARY
Upper gastrointestinal endoscopy is a safe procedure with a low morbidity and mortality. With its increasing use, previously described complications (Schiller and Prout, 1976), of which those of a respiratory nature are most prominent, could become more apparent. Cyanosis is not infrequently noted during the examination, respiratory depression or apnoea may result from the sedatives administered
upper gastrointestinal endoscopy and gave fully informed consent for the study were investigated. One hour before the procedure, the FEV1 and FVC were measured, using a wedge bellows spirometer (Vitalograph Ltd.). Symptoms and past history of cardiovascular disease, respiratory disease, and' smoking habit were recorded. The results of tho previous chest radiograph and ECG were noted. Fifty patients (group 1) received 15 puffs of metered lignocaine throat spray (Xylocaine spray, Astra Chemicals Ltd.: one puff equals 10 mg lignocaine), followed by intravenous sedation with diazepam. A further 15 patients (group 2) were subjected to endoscopy, having only 15 puffs of throat spray and no premedication of intravenous sedation. A radial arterial puncture was made for a baseline arterial blood gas tension analysis, after which the patient's throat was sprayed with lignocaine and in group 1 intravenous diazepam administered until drowiness was achieved, the amount; given being noted. Endoscopy was performed with an, Olympus GIFK endoscope and a second arterial' blood sample was taken when the stomach and' duodenum had been examined, but with the endo-scope still remaining in situ. This occurred at a mean time of 12 minutes. Blood gases were measured using. a Radiometer ABL 1 automatic blood gas analyser. The patients were monitored electrocardio-graphically throughout the procedure and any arrhythmias noted. In addition, the efficacy of' sedation and the occurrence of cyanosis were noted.
(Dundee and Haslett, 1970), and pulmonary aspiration can occur in up to 29 % of procedures (Prout and Metreweli, 1972). Electrocardiographic abnormalities have been demonstrated in a variable proportion of examinations (De Masi and Akdamar, 1969; Sturges and Krone, 1973; Py6rala et a!., 1973; Fujita and Kumura, 1975), particularly in the presence of ischaemic heart disease, and it has been suggested that hypoxia may be a contributory factor. Although blood gas tensions have been shown to change during fibreoptic bronchoscopy (Salisbury et al., 1975), to our knowledge such changes have not been studied during upper gastrointestinal endoscopy. The purpose of this study was to measure the change, if any, in arterial blood gas tensions before and during endoscopy and to relate any changes observed to the administration of sedative and occurrence of cardiac arrhythmias. Forced expiratory volume in one second (FEV1) and forced vital capacity (FVC) were measured beforehand to assess their predictive value for patients at risk. Methods
Results
Sixty-five consecutive patients who presented for Received for publication 29 June 1976
Table 1 gives the details of the patients in the studyand Table 2 shows the results for PaO2 and PaCO2,. 797
798
P. J. Whorwell, C. L. Smith, and K. J. Foster
Table 1 Details ofpatients Group Male
Female
no. % no.
1
I1I
35 70 15 11 73 4
Y/.
30 27
Age(yr) Mean
Mean Hb Range
(yr) 55 6 55 3
18-82
26-76
(gldl)
12-8 13-4
Diagnosis Ulcer
Gastritis
Gast. no. %
Duod. no. Y.
no.
7 3
4 1
8 3
14 20
8 7
Duodenitis Carcinoma Normal
Other
Y.
no.
16
7 3
20
% 14 20
no.
%
no.
%
no.
%
4 0
8
17 4
34
26
3 1
6 7
Table 2 Arterial PaO2 and PaCO2 before and during endoscopy with and without diazepam Group
No.
Initial PaO, (KPa)
PaOQ during endoscopy (KPa)
Percentage difference
Mean ± SE
Mean i SE
Mean i SE
10-62
I
50
12-32 ± 0-24
(Diazepam) II (No diazepam)
15
12-34
1
49
Initial PaCO, 4-87 i 0-08
15
4-63 i 011
0 53
0-23
13-39
1-42*
10-36 t 0-42
14-78
3-69*
PaCO2 during endoscopy
Percentage difference 7-08 1-40*
520 i 009
(Diazepam) II (No diazepam)
4-49
0-13
3-63
2-28
*P < 0-001.
obtained before and during the examination in both groups of subjects. In group 1 the mean PaO2 of 10-6 + 0-23 KPa found during the examination is significantly lower than the mean initial PaO2 of 12-3 + 0-24 KPa (p < 0-001). The mean percentage difference in PaO2 between the initial and subsequent readings of 13-4% is highly significant (p < 0-001). Similar results for PaO2 were obtained in group 2, where no diazepam was used. The number of patients examined in this group was limited to 15 as the study was continued only until it had been confirmed statistically that there was no difference, in respect of the fall of PaO2, between the two groups. In group 1 there was a significant increase in the PaCO2 during endoscopy, contrasting with a marginal decrease in PaCO2 in group 2 where no diazepam was administered. The Figure shows the least squares regression line for both groups for the correlation between the percentage change in PaO2 and the starting value. There was no significant difference in the separate regression lines for the two groups, and thus the combined regression line, which is highly significant (p < 0-01), is shown. It is clear that in all but two subjects there was a fall from the initial PaO2, regardless of the starting value, although there was a tendency for those with the highest values to fall by a greater extent. Although the FEV1 correlated with the initial PaO2 (P < 0-05),
neither the FEV1 nor the FVC were of predictive value for the percentage fall in PaO2. After the exclusion of those patients with preexisting arrhythmias, in group 1 arrhythmias occurred in seven patients (14 %), of whom two gave a history of ischaemic heart disease. In group 2, only one patient, who had ischaemic heart disease, had an arrhythmia. All arrhythmias except one were multiple ventricular ectopics, the exception being atrial fibrillation. All reverted spontaneously after the procedure. The occurrence of arrhythmias was not related to the degree of fall in the PaO2. There was no correlation between the dose of diazepam administered or depth of sedation produced and the percentage fall in oxygen tension. Discussion
This study clearly shows that there is a marked fall in the arterial PaO2 during uncomplicated upper gastrointestinal endoscopy. Although the mean PaO2 during endoscopy was only just below the lower limit of normal, in many subjects it fell significantly further, the lowest recorded value being 5-9 KPa. The observation that there was an equally notable fall in PaO2 when no diazepam was administered indicates that this effect is not due to the diazepam. It therefore seems probable that it is either the presence
Arterial blood gas tensions during upper gastrointestinal endoscopy * With
0to °2 Change 50
799
Diazepam
* Without Diazepam
40 _
30
a
Figure The percentage change in PaO2 plotted against the starting value, with the least squares t regression line for the combined data, is shown. The curved lines show the 95 % confidence limits for the regression line.
a
20
10
8 -10
10
12
14
*
16
18 Before 02
Endoscopy
-20
of the endoscope or the lignocaine throat spray, or a References combination of the two, which is responsible for this finding. It has been previously reported that ligno- De Masi, C. J., and Akdamar, K. (1969). Electrocardicgraphy during upper gastrointestinal endoscopy. Gastrointestinal caine throat spray may affect blood gas tensions Endoscopy, 16,33-34. (Salisbury et al., 1975). However, diazepam may Dundee, J. W., and Haslett, W. H. K. (1970). The benzocontribute to some extent in that the mean PaCO2 diazepines. British Journal of Anaesthetics, 42, 217-234. rose only when diazepam was administered. Dunn, G. D., Kubin, R. H., Laing, R. R., Sisk, C. W., and Klatz, A. P. (1970). Double-blind study of endoscopy preAlthough hypoxia may contribute to the generamedications. Gastrointestinal Endoscopy, 16, 229-230. tion of cardiac arrhythmias, in this study no clear Fujita, R., and Kumura, F. (1975). Arrythmias and ischemic relationship between hypoxia and cardiac rhythm changes of the heart induced by gastric endoscopic abnormalities could be shown in our patients. procedures. American Journal of Gastroenterology, 64, 4448. Although upper gastrointestinal endoscopy is a G. R., Kehoe, E. L., Friedman, E., and Belber, J. safe and diagnostically invaluable procedure, the Mayes, (1970). Pre-endoscopic medication: parenteral diazepam present study indicates that circumspection should used adjunctively. Gastrointestinal Endoscopy, 16, 187-193. be used in submitting to endoscopy those patients Prout, B. J., and Metreweli, C. (1972). Pulmonary aspiration after fibre-endoscopy of the upper gastrointestinal tract.. with respiratory disease, and the previously advoBritish Medical Journal, 4, 269-271. cated use of potential respiratory depressants as pre- Prout, B. J., and Schiller, K. F. R. (1976). In Modern Topics medication (Dunn et al., 1970; Mayes et al., 1970; in Gastrointestinal Endoscopy, pp. 73-81. Edited by K. F. R. Prout and Schiller, 1976) is to be discouraged. Schiller and P. R. Salmon. Heinemann: London. We are grateful to Mr D. Carpenter for the blood gas analyses and to Mr J. Alexander for his assistance with the statistical analysis.
Pyorala, K., Salmi, H. J., Jussila, J., and Heikkila, J. (1973).. Electrocardiographic changes during gastroscopy. Endoscopy, 5, 186-193. Salisbury, B. G., Metzger, L. F., Altose, M. D., Stanley, N. N., and Cherniack, N. S. (1975). Effect of fibreoptic
P. J. Whorwell, C. L. Smith, and K. J. Foster
800 bronchoscopy on respiratory performance in patients with chronic airways obstruction Thorax, 30, 441-445. Schiller, K. F. R., and Prout, B. J. (1976). In Modern Topics in Gastrointestinal Endoscopy, pp. 147-165. Edited by K. F. R. Schiller and P. R. Salmon. Heinemann: London. Sturges, H. F., and Krone, C. L. (1973). Cardiovascular stress
of peroral gastrointestinal endoscopy. Gastrointestinal Endoscopy, 19, 119-122. Requests for reprints: Dr. C. L. Smith, Professorial Medical Unit, Southampton General Hospital, Tremona Road, Southampton S09 4XY.
The September 1976 Issue THE SEPTEMBER 1976 ISSUE CONTAINS THE FOLLOWING PAPERS
Histopathology of cell mediated immune reaction in mouse colon-allograft rejection R. J. HOLDEN AND ANNE FERGUSON
Ischaemic colitis in the experimental animal I Comparison of the effects of acute and subacute vascular occlusion
J. G. W. MAITHEWS AND T. G. PARKS
Ischaemic colitis in the experimental animal U Role of hypovolaemia in the production of the disease
Idiopathic cholestasis of pregnancy in a large kindred
HUMBERTO REYES, J. RIBALTA, AND M. GONZALEZ-CERON
Autoimmune reaction to a liver specific membrane antigen during acute viral hepatitis A. M. G. COCHRANE, A. MOUSSOUROS, A. SMITH, A. D. THOMSON, A. L. W. F. EDDLESTON, AND ROGER WILLIAMS
Gastritis duodenitis, and circulating levels of gastrin in duodenal ulcer before and after vagotomy D. D.
J. G. W. MATrHEWS AND T. G. PARKS
MEIKLE, K. B. TAYLOR, S. C. TRUELOVE, AND R. WHITEHEAD
Exocrine pancreatic function in juvenile-onset dia-
Tranexamic acid and upper gastrointestinal haemorrhage-a double blind-trial J. C. BIGGS, T. B. HUGH,
betes mellitus B. M. FRIER, J. H. B. SAUNDERS, K. G. WORMSLEY, AND I. A. D. BOUCHIER
Stenosis of the colon in acute pancreatitis w. s. J. MAIR, M. J. MCMAHON, AND J. C. GOLIGHER
AND A. J. DODDS
Transient paraproteinaemia in a patient with coeliac disease A. S. PENA, JANNY V. NIEUWKOOP, H. R. E. SCHUIT, W. TH. J. M. HEKKENS, AND A. J. CH. HAEX
Technique Peroral small-intestinal biopsy: experience Ferritin in Crohn's disease tissue: detection by with the hydraulic multiple biopsy instrument in electron microscopy P. J. WHORWELL, R. C. routine clinical practice BRIAN B. scolT AND M. S. BALDWIN, AND R. WRIGHT
LOSOWSKY
Hypergastrinaemia in cirrhosis of liver sHu KUM
Notes and activities Notes on books
LAM
Copies are still available and may be obtained from the PUBLISHING MANAGER, BRMSH MEDICAL ASSOCIATION, TAVISTOCK SQUARE, LONDON, WC1H 9JR, price £2-00, including postage
